医学
列线图
细菌性脑膜炎
脑膜炎
重症监护医学
儿科
内科学
新生儿败血症
梅德林
预测模型
作者
Min Yu,Yiping Fang,Xiaofeng Yang
标识
DOI:10.21037/tp-2026-0206
摘要
Background: Neonatal bacterial meningitis (NBM) remains a life-threatening condition with considerable mortality and long-term neurodevelopmental impairment. Early prognostic assessment is challenging, and prediction models tailored to neonates are limited. This study aimed to develop and internally validate a prognostic nomogram for predicting poor outcomes in neonates with bacterial meningitis. Methods: A retrospective cohort study was conducted including 214 neonates diagnosed with bacterial meningitis between 2011 and 2024. Independent predictors of poor outcome were identified using Firth penalized logistic regression to reduce small-sample bias. A prognostic nomogram was developed and internally validated using bootstrap resampling to assess model performance. Results: Among the 214 neonates, 83 (38.8%) experienced poor outcomes. Lower gestational age, seizures, elevated cerebrospinal fluid (CSF) protein, and reduced CSF glucose were independently associated with poor prognosis. The model demonstrated good discrimination, with an area under the curve (AUC) of 0.79 [95% confidence interval (CI): 0.72-0.85] and an optimism-corrected C-index of 0.77 after bootstrap validation. Calibration analysis showed good agreement between predicted and observed probabilities. Decision curve analysis suggested net clinical benefit across threshold probabilities ranging from 0.05 to 0.65. Conclusions: This model integrates inflammatory biomarkers reflecting central nervous system injury with clinical parameters, providing a potential bridge between pathophysiological mechanisms and bedside risk stratification. External validation in independent cohorts is warranted prior to routine clinical application.
科研通智能强力驱动
Strongly Powered by AbleSci AI